SURVIVAL OF THE COMMONS: MOUNTING CHALLENGES AND NEW REALITIES THE INTERNATIONAL ASSOCIATION FOR THE STUDY OF COMMON PROPERTY June 19-23, 2006. Bali, Indonesia ______

Decentralisation, ‘Recentralisation’ and Irrigated Farmers in Torsten Rødel Berg Centre for Development and International Relations, Aalborg University. [email protected]

Abstract

This paper examines the meanings of privatisation and political decentralisation as complementary approaches to the transfer of power, on irrigated communities in Nepal. The introduction of democracy in Nepal in 1990 coincided with a shared donor agenda emphasising private sector participation in the management of natural resources. The notion of the ‘local resource management institution’ as an alternative to government bureaucracy was pervasive; this led to its resurrection in rural development policy and to a central role for community based-management in Nepal’s rural development strategy. Parallel to this, democratic Nepal had embarked upon a process of political decentralisation involving the setting up of democratically elected local government bodies. The combined process resulted in diverse rural institutional landscapes, connected formally through legal recognition of user’s groups and NGOs, and informally through the dynamics of political economy, with districts as local political arenas. As an outcome of an ongoing political crisis in Nepal the political decentralisation process came to a halt when local bodies were abolished by the central government in 2002; likewise, restrictions on civil liberties and Maoist attempts at controlling civil society continue to reduce the room for manoeuvre for community-based institutions.

This trajectory offers an opportunity for exploring the experience of deconcentration in a ‘with’ and ‘without’ perspective. This is done through narrative analysis of the institutional context of the democratic era, and its meanings in terms of livelihoods among irrigated farmers in the hills of West-central Nepal. With irrigation institutions as central elements in the agricultural development policies of the 1990s, and as a relatively better-off segment of rural society in a relatively well-endowed area of Nepal, these farmers were in a good position to reap benefits associated with the institutional plurality of the democratic era. Hence their plight during the present process of ‘recentralisation’ is particularly relevant to evaluations of the overall decentralization approach in a with and without perspective

The paper finds that the institutional plurality of the democratic era was important to many farmers and communities. Livelihoods were played out in the context of both community- based (including common property institutions) and government institutions, connected through a combination of legal structures and ‘power’ in socio-economic relations. Cooperatives, irrigation and forestry user groups brought economic benefits and provided

1 community leverage vis-à-vis government bodies. In many cases Village Development Committees were instrumental in connecting these community institutions – which were often used as platforms for local elites to gain political power - with the surrounding political landscape. The paper concludes by suggesting that analysis of the decentralization experience in terms of its livelihood meanings should go beyond notions of ‘ private’ and ‘political’ spheres to include analysis of ‘political space’. This involves looking at how farmers navigate within the institutional landscape, their organizing practices, resultant service delivery and resource allocation in specific socio-economic and cultural contexts.

The paper is based on a research project, which investigates socio-institutional change in irrigated communities in mountain and hill districts in Nepal.

1. Background

The rural development trajectory of Nepal has moved in conflicting directions since democracy was restored in 1990. In stark contrast to previous centralist policies, democratic policies involved transfers of power from central bureaucracies through political decentralisation with the setting up elected local multi-purpose bodies (Village Development Committees and District Development Committees) and facilitated widespread ‘privatisation’ (Ribot 2004) decentralisation, i.e. single-purpose user groups, committees etc. as part of promoting civil society and pluralism. This process resulted in the emergence of diverse institutional landscapes and the articulation of a vast plurality of interests (Berg 2003). Parallel to this, the rural development planning frameworks emphasised not only local participation but also an economic model based on agriculture as the engine for growth through ‘growth-linkages’. Yet, rural livelihoods have, particularly in the past decade, diversified away from agriculture, and local economies have become more strongly associated with employment opportunities in India and, increasingly, global labour markets.

Newly democratic Nepal of the early 1990s had embarked on not only decentralisation reform, but also policies of economic liberalisation, and the notion of the local resource management institution as an alternative to management of (local) resources by central government bureaucracies was pervasive. Not least in the irrigation sector (see Vermillion 2001) which had experienced some 4 decades of centralist ‘development’. However, community managed irrigation was not new to Nepal: Estimations of the number of irrigation systems range from 20.000 (Pradhan 1989) to 100.000 (GTZ 1991), the vast proportion of which had evaded government interventions and remained locally managed. The process of gradually putting the local resource management institution in irrigation back on the agenda, to the extent that it came to occupy a significant role in the rural development strategies and policy frameworks of Nepal1 was helped along by a series of influential studies that rediscovered the historical role of indigenous irrigation.

Along the lines of common property theory approaches these sought to define factors of success, organisational form, functions and logics (see Pradhan 1989, Yoder 1994, Ostrom, Benjamin et al. 1992, Rana 1992, KC and Pradhan 1993, Pradhan and Yoder 1990, ILO 1995,

1 Including the current rural development framework (to 2015) that has been prescribed by the Agricultural Perspective Plan of 1995. The plan emphasizes reliance on agriculture as the engine for growth through a ‘growth linkages model’; obstacles to growth are considered as technological, with access roads, fertiliser, irrigation, and research as key policy foci (see Cameron 1998), along with irrigation institutions (NPC/APPROSC 1995).

2 Pradhan 2003). Evidence of community managed irrigation systems performing better than government-managed systems (Ostrom 2002) constituted another determinant with respect to assigning priority to community-based irrigation. This lead to legal reforms that took away power from traditional line agencies and allowed ‘participation’ by farmers, mainly in the construction of irrigation schemes, and in terms of representation at district level in the emerging decentralised elected local government structures. Social mobilisation activities became key elements in irrigation development, and eventually ownership of irrigation structures came to rest with communities whose irrigation institutions became responsible for operation and maintenance2.

In sum, farmer’s irrigation associations and other prominent single-purpose user groups such as farmer’s cooperatives and forestry groups became connected to a surrounding institutional landscape where decentralisation policies had enabled their legal recognition as user groups and thus (in principle) eligibility for funding through local administrative bodies. The process gained momentum with the introduction of the 1999 Local Self-Governance Act and moved further beyond mere territorial decentralisation to substantially include devolution of decision-making powers and the building up of capacity for planning and implementation in local, elected bodies. However, amid growing political turmoil, involving escalation of the conflict between the armed forces and Maoist insurgents, the Village Development Committees and District Development Committees, were abolished by the central government in 2002. At district level, the latter is now represented by the Chief District Officer (the Magistrate), the Local Development Officer, the police and army, as well as the traditional line agencies, all of which are appointed by and , as in the pre-democracy days, accountable only to the central level.

2. Conceptual and Analytical Issues

Until a few decades ago dominant rural development discourse, based on modernisation- inspired neoclassical tradition did not attach serious importance to institutions other than the market. That changed with the emergence of neo-institutionalism, the ‘post-structural turn’ (Blaikie and Coppard 1998) and the emergence of “spaces in which grassroots action flourished from the mid-1980s onwards…created in some measure by the backing off of big government from heavy-handed involvement in the rural economy” (Ellis and Biggs 2001:443). In Nepal and other developing countries the transfers of powers from large bureaucracies to non-state entities – non-government organisations (NGOs), user groups etc - often in the name of decentralisation (Ribot 2004) - were attuned to the liberal environment promoted by international donors (Start 2001). Theoretically, the focus on the ‘local’ was accommodated within the neo-classical rural development tradition because of a focus on behaviour and choice at individual and group levels, as well as a strong degree of agency

2 This marked a departure from previous centralised project implementation practice that involved only line agencies and a hierarchy of contractors, made possible as a result of amendments to His Majesty’s Government’s Financial Rules that assigned “priority to user’s committees formed by local users for the construction of projects to be implemented in rural areas up to a cost ceiling of Nrs. 10 Lakh” (Nrs. one million) (Ch. 42, no. 46, Nepal Gazette part 3, 01 March 1993: 9). These amendments followed wording in the Eighth Plan (1992-1997) emphasising that “rural communities will be encouraged to organise user groups for building community infrastructure and to organise beneficiary groups in order to strengthen receiving mechanisms at the grass roots level. All projects within the technical and managerial competence of the local people will be implemented by local user groups with assistance from local and district committees” (NPC 1992: 41)

3 orientation in the central body of theory associated with decentralised natural resource management, viz. common property theory3.

The theoretical underpinnings of this approach are based on the strand of (new) institutional economics that deals with collective action in relation to public or collective goods and common property resources (Nabli and Nugent 1989). Within this rational choice framework that stresses the micro-foundations of institutions (Scott 2001), it has been demonstrated that the presence of specific rules and incentives may overcome the problem of free riding, and that local, natural resource management institutions—based on individual economic rationales as determinants for investment in collective action—may be efficient ways of organising economic activity (Ostrom 1990). The identification of enabling factors for collective action (including social homogeneity and supportive traditions) led to the elaboration of design principles (Ostrom 1990, Vermillion 2001) and the persuasive notion that appropriate local institutions might be engineered.

While not opposing the idea that commons do work under certain circumstances, it has been argued that the tendency within common property research to explain the workings of e.g. irrigation institutions at rather internal levels limits its capacity for generalisation (Dayton- Johnson 2001) and the capturing of change (Moench 2002), social, political and economic. It has also been argued that the tendency to analyse communities and associated common property resource regime management institutions as stable entities in which people are seen to have incentives to collectively manage key resources (i.e. agriculture) is increasingly problematic (Sadeque 1999, Moench 2002) if agriculture is no longer as central to rural economies and to people’s livelihoods as it used to be.

This calls for a less restrictive take on institutions that goes beyond assuming that common endeavours are “simply the aggregated outcome of the effective agencies and interests of individuals” (Long 2001: 57) but which to a greater extent acknowledges that institutions are not outside the social relations that shape them (Berry 1989) or outside the influence of external economic and political forces, yet recognises the role of agency. Livelihoods analysis may be pointed towards as an approach to accommodating both structure and agency. In this context, Leach, Mearns et al 1997 suggest that the entitlement concept (on which livelihoods analysis is based and which refers to the range of possibilities that people can have), is about gaining access to resources and that formal legal institutional mechanisms as a means of gaining such access, are part of the entitlement framework. Thus the ‘investments’ that people make in collective action are part of livelihood construction and institutional relations and practices influence the ways in which people are able to transform their assets - natural, physical, human, financial and social – (see Ellis 2000) as “livelihoods and livelihood strategies are crafted and played out within the context of varying sets or matrices of institutions” (Bingen 2000:9).

While the above theoretical baggage (and the livelihoods approach in particular), is central to understanding the meanings of institutions in the irrigated communities that are under investigation, this paper’s focus on the democratic trajectory of Nepal’s recent past calls for

3 While, admittedly, too general a notion and an oversimplification, common property theory refers to the academic discourse that started out in reaction to Hardin’s (1968) pessimistic views on the viability of common ownership of natural resources. Currently, the theories and concepts for solving collective action problems in rural development are applied in relation to management and governance of the commons, referring to a “diversity of resources or facilities as well as to property institutions that involve some aspect of joint ownership or access” (Dietz, Dolzak et al 2002:18).

4 additional conceptual inputs. In recent years the debates on the role of common property regimes have been running in parallel with debates on democratisation and (‘good’) governance in natural resource management. With the livelihoods approach in mind, this paper engages to some extent with this related debate where foci are on the public and private spheres of institutions and include equity aspects (Meynen and Doornbus 2004), democratic (elected bodies) versus ‘privatisation’ (user groups) decentralisation (Ribot 2004), as well as the effectiveness of single-purpose institutions (user groups) versus multi-purpose institutions, i.e. elected councils with wide portfolios (Manor 2004).

However, while the concepts of single-purpose and multi-purpose institutions are useful for understanding institutional characteristics in the current context, the experience of irrigated farmers during processes of decentralisation and ‘recentralisation’ suggest that private/public dichotomies are not necessarily useful in terms of understanding livelihoods and institutional trajectories. Therefore, it is posited, the broad concept of political space (Webster and Engberg-Pedersen 2002) is a useful additional theoretical tool. It is understood as “the types and range of possibilities present for pursuing poverty reduction by the poor or on behalf of the poor by local organisations” (Webster and Engberg-Pedersen 2002:8), and implies that ‘the poor’ straddle private/public spheres and navigate within the institutional landscape regardless of private/public categories. Political space is constituted by three main components: Institutional channels for accessing policy formulation and implementation, political discourses of poverty and poverty reduction, and; social and political practices that may influence policy agendas and actual implementation (Webster and Engberg-Pedersen 2002). Political space and the dynamics that it entails is “more likely when there has been a transition towards some form of democracy” (Webster and Engberg-Pedersen 2002:11) as was the case in Nepal from the early 1990s to the early 2000s.

3. Methodological Issues

Relationships between livelihoods and institutional change are under investigation in 3 irrigated communities in the mountain district of Mustang, and 6 communities in the hill districts of Myagdi, Baglung and Parbat in the Dhaulagiri Zone in Western Nepal. These communities, representing some 1000 households (or 5000+ individuals) have been selected for investigation because of the availability of data that enables comparison of socio- economic and institutional circumstances in 1992-19954 with those of 2004. The re-study has followed largely similar methodologies and thematic areas as previous studies, and the 1992- 1995 and the 2004 ‘snapshots’ contain data on population, landholding size, and parameters of livelihood diversification such as income, employment, migration, as well as data on yields, cropping patterns and cropping intensities. Institutional aspects include investigations into parameters of cooperation and performance such as organisational structure, operation and management practices, including labour inputs and the basis for labour contribution, as well as water supply, distribution and allocation. Standard sampling procedures and a combination of quantitative and qualitative data collection and analytical techniques have been applied. This paper relies predominantly on findings related to qualitative investigations, i.e. on interviews with ‘key informants’ and other informants that have been

4 As part of the ILOs Dhaulagiri Irrigation Development Project’s monitoring and evaluation activities socio-economic baseline and key- effects studies were conducted in 15 irrigation systems in all four districts on an annual basis in the period from 1992 to 1997. The author was in charge of the design and implementation of surveys from 1992-1995

5 selected on the basis of principles associated with maximum variation sampling. The overall research approach is one of seeking explanations through an inductive approach.

4. The Dhaulagiri Area

The 9 gravity-irrigated systems under study are located along the upper Kali Gandaki River5, concentrated along a 75-km transect from in the south, through Baglung and Myagdi Districts, northwards to . Between Myagdi and Mustang, the Annapurna and Dhaulagiri Massifs bisect the area, which changes from the hill ecological region to the mountain ecological region. Elevation ranges from 800-1200 meters above sea level in the 6 irrigation systems6 that are located in the hills, under subtropical conditions with monsoon rains, with paddy, wheat, maize and vegetables as the main irrigated crops. The proportion of agricultural area that is irrigated in the hill districts is around 10%. The 3 irrigation systems7 of Mustang District, ranging from 2800 – 3500 meters above the sea have a temperate climate; located in the rain-shadow of the Himalayas conditions are arid, subsequently the irrigated proportion of the agricultural area is very high (83%) and non- irrigated agriculture largely futile. Buckwheat, naked barley, maize and apples constitute the main irrigated crops.

The Dhaulagiri Zone is characterised by what has for centuries been a major trade route linking mountains and hills with the plains of Nepal and India. Both irrigated (khet) and unirrigated (bari) terraces surround the hill portion of the area, which is predominantly inhabited by Hindu Parbatiyas8 residing in relatively scattered settlements. Further north, Hindu culture gradually gives way to Buddhist culture: Emerging from the deep gorge between the Annapurna and Dhaulagiri ranges, the physical, cultural and institutional landscape is entirely different from that just a few days walk to the south. Tibeto-Burman Thakalis and Bhotiyas reside in nuclear settlements (ranging from 2800-3500 meters above sea level) surrounded by green, irrigated fields, appearing as oases in the desert. Throughout the 1990s a road was gradually constructed from the south along the trade route, every extension accompanied by an increase in the economic importance of the roadhead location. Road construction has also been in progress from north to south for some years.

Seddon and Adhikari observe, in the context of livelihoods security in Nepal that

“There is a high degree of heterogeneity at all levels – agro-ecological (i.e. mountains, hills and teria), regional (western, eastern etc), district, Village Development Committee, hamlet, household and individual” (2003: no page number)

In the context of the Dhaulagiri Zone variation and the political and institutional complexity that follows is particularly pronounced in the hills where also population densities are much higher than in Mustang District. It is also in the hills that development in the shape of road infrastructure and market access coincides with the area where the irrigated communities of

5 The Gandaki Basin stretches from the Dhaulagiri Range to the Langtang Range. The Kali Gandaki River originates in Northern Mustang, on the Tibetan Plateau. 6 Amalachaur, Arjewa, Kurgha, Pakuwa, Lampata, Pipalbot 7 Thin, Tiri and Khinga 8 Parbatiyas (literally ‘people of the mountains’) is a term denoting hill residing caste hindus whose ancestors migrated into the subcontinent from the west, reaching what is now western Nepal during the first millennium AD (Whelpton 2005).

6 this study are concentrated. As irrigated communities, at low altitudes, close to main trading arteries these are the types of communities that historically tend to be relatively prosperous. These initial endowments in combination with road access and a distinct roadside bias meant that the spoils of the democratic era (information, NGOs, projects) penetrated to these communities to a much larger extent than was the case in the hinterlands.

5. Livelihood Contexts

Both hills and mountain communities and households are increasingly integrated into the wider economy, both at the local levels through commercial channels and at more global levels through integration into global labour markets. However, the mountain communities, regardless of their relative isolation are more deeply integrated than the hill communities. The nearest roadhead is located 2 days from the irrigated communities in Mustang. Sale of agricultural products is therefore largely intra-district with Jomsom, the district headquarters, representing substantial demand (see Vinding 1984) owing to the presence of army and government personnel, as well as its role as a tourist centre and transport hub. North-south trade in livestock (yaks) is a major source of agricultural income in all three communities. Apples play some role in terms of extra-district agricultural trade, both fresh and processed into brandy, but transportation costs remain a hindrance to the expansion of exports. Staples grown under irrigated conditions are generally consumed within farming households, either by humans or livestock. The role of tourism has increased over the past decade and plays a role with respect to demand for vegetables, and some families gain income from the operation of hotels and shops.

However, traditional, seasonal trade involving seasonal migration is the major source of income in these mountain communities. This typically involves husbands and sons from most households travelling south for 3–4 months every winter, selling livestock en route in Nepal, and sweaters, and (other) Chinese goods, as well as herbs in India. Livestock and consumer items are often purchased on the return journey. In addition, the prevalence of more permanent migration is high; some 60% of all households have migrated family members. While most emigrants hold a variety of jobs in Nepal (including government service, contracting and hotel business) or are students, a substantial proportion work abroad, predominantly in the relatively lucrative labour markets of Japan and South-east Asia. Migration to the latter destinations has increased substantially over the past decade. Remittances from emigrated family members are important for a minority of households only. Average, annual household cash incomes at over US$ 2000 are some 2.5 times higher than those in the Hills and have grown, on average, by almost 50% per annum over the past decade.

In the hill communities agriculture has diversified considerably over the past decade to include – in addition to conventional grain crops – coffee, bananas and vegetables, the latter often grown in irrigation command areas. All communities are within relatively close proximity of the main road or its feeder roads (ranging from 1 hour to ½ a days walk) and agricultural produce (so-called off-season vegetables) is sold in the growing markets of the district headquarters and beyond. Unlike their mountain peers, hill farmers tend not to engage in commercial activities other than the sale of agricultural products. Yet non-farm income constitutes the main source of income in these communities, most of it derived from remittances from migrated family members. On average, 40 % of households have one or more members who work in other parts of Nepal, in India or beyond. The growth in demand

7 for agricultural products combined with overseas migration means that household incomes – averaging US$ 800 per annum - have increased by some 50% over the past decade. The majority of migration in the hills is permanent and typically involves both husbands and sons, with a tendency for mothers, daughters and the elderly to stay behind.

6. Characteristics and Performance of Irrigation Institutions

Governance arrangements differ substantially between mountain and hill communities. In all three Mustang communities the irrigation system, along with other common property resources, is governed through century old multi-purpose village councils whose responsibility is to organise public works, enforce regulations, manage funds, solve conflicts, summon meetings, and impose fines with respect to all natural resource management, including irrigation, forestry, grazing, livestock and infrastructure. These councils are lead by headmen (mukhyia) elected at general assemblies for 2-year duties, by all residents. Observers (Furer-Haimendorf 1975, Messerschmidt 1995, Parajuli and Sharma 2000) agree that village councils with consensual leadership and high accountability, along with strong cooperation from villagers in common property management, are critical factors for survival under harsh physical conditions. Other resource mobilisation dynamics associated with the ‘communities of trust’ and shared interests (Seddon 1979) that characterise the sparsely populated area include rotating credit associations (dhikurs) (Furer-Haimendorf 1975, Messerschmidt 1995).

A variety of fragmented cooperative systems for the management of natural resources, particularly forests and irrigation exist in the hill systems (see Pradhan 1989, Yoder 1994), but unlike in the mountains these consist in their present form of single purpose committees, representing various interest groups. In the case of irrigation, formal bodies responsible for canal governance were only institutionalised in the early 1990s, as part of physical and institutional rehabilitation project efforts. The intensity of operation and maintenance efforts vary, ranging from the voluntary assumption of responsibilities on the part of individual families to the temporary employment of caretakers as and when required, whose main task is to oversee water allocation on a rotational basis. Labour contributions for maintenance (based on landholding size) are voluntary in practice, and abstention is commonplace. While provisions for the fining of absentees do exist, very few cases of enforcement are reported.

This situation stands in contrast to the more structured mountain systems, where – in addition to the mukhya, permanent caretakers (katuwal), whose renumeration is sometimes monetarised, but most often in kind, are employed by the village council. These katuwal are responsible for routine maintenance and for the organisation of both regular and emergency repair works. Irrespective of landholding size each household has to send at least one adult person for such works, and abstention (reportedly rare) involves the levying of fines (of 3-4 US$ per day). In the larger mountain systems allocation is also rotational, but in the smaller mountain systems water is allocated and distributed to individual farmers based on ‘owned’ water shares. Small farmers get fewer shares per irrigation cycle than do larger farmers. The shares correspond to specific dates decided upon by means of a lottery. On the given dates, it is the responsibility of the farmer whose turn it is, to distribute water to his/her own field.

Labour inputs per household for both routine and emergency works have been reduced by half since 1992-1995. In the mountains the reduction reflects reduced maintenance requirements following physical rehabilitation. In the hills this is not the case; in at least 4 of the systems landslides and floods have damaged structures to the extent that – as assessed by farmers –

8 labour requirements are no less now than prior to rehabilitation. Farmers in these communities assign labour shortages as the reason for reduced inputs. Linked to this, while 65% of the hill farmers expressed satisfaction with water quantity in 1994, only 33% of farmers did so in 2004. In the mountains, the corresponding figures are 72% and 65%. While some 71% of hill farmers also thought that they received ‘fair’ shares of water in 1994, that figure was down to 24% in 2004. In the mountains the corresponding figures were 100% in 1994 and 87% in 2004. The picture of mounting institutional problems in the hills is reinforced by reduced average yields of the main irrigated crops: Paddy yields have dropped in 4 out of 6 hill systems, from 3.6 tonnes per hectare in 1994/1995 to 3.3 in 2004. Indications are that yields of wheat have been reduced substantially as well (by up to 50% in some cases) while maize may have dropped by 25%. It should be noted that the hill systems without reduced paddy yields are also the least migration-afflicted communities. In the mountain systems the data suggests stagnating rather than falling yields. Similarly, cropping intensities in the mountains (at an average of 172 in 2004) are fairly stable, while in the hills they have fallen from an average of 252 in 1994-1995 to 239 in 2004.

7. Institutional Trajectories

The above mentioned characteristics and performance need to be understood, not just in terms of organisational form, but also in the political context of the past decades. The single purpose institutions of the hills became formalised, legal entities as part of the irrigation rehabilitation project on which the ‘historical’ data of this research is based, along common property theory principles, and in line with government (and donor) efforts to decentralise power. The standard formula involved inputs by social mobilisers, recorded membership of Farmer’s Irrigation Institutions, elected committees, adherence to byelaws and registration with local authorities (the District Development Committees and the District Irrigation Offices via the Village Development Committees).

However, the irrigation committees have disappeared in two of the six hill communities, seemingly as a result of the migration of members. Existing committees show signs of coping problems and members point to the severance of links with local authorities, along with the civil war as the main reasons. Renewal of registration ceased since the start of the 2000s, as political life became increasingly problematic, not least for elite village politicians who had used committees as platforms for advancement. Additionally, local authorities had become increasingly unresponsive to local institutions in the face of the growing political crisis, triggered by increased Maoist influence in rural areas, that culminated in the abolishment of local authorities in 2002.

In the mountains some degree of integration of village councils with government administrative systems was a feature of both the (pre-1990) panchayat structures (when Mukhyas or village heads where appointed by the central authority), and the post 1990 democratic decentralisation structures when Village Development Committees and District Development Committees were indeed set up, but performed predominantly as practical facades. Farmers generally recollect the interfaces with both the pre-democratic and the democratic structures of the national political system as negative (except for the annual grant allocation) because nepotism and cooption/corruption of traditional leadership was seen to divert attention away from local concerns and the ingrained democratic features9 of the

9 Vinding (1998) describes the traditional political system as essentially democratic because of the widespread practices of voting.

9 traditional political system. These features were badly understood by external actors. As an example of this, the efforts at institutional crafting with respect to irrigation institutions in the 1990s completely ignored the existence of traditional governance structures and followed the prescriptions of the hills. Today, hardly any trace of these externally imposed institutional structures remains.

However, the democratic period did leave its mark: The abolishment of local authorities in 2002 and the absence of annual grants (that had been controlled not by the Village Development Committees but the traditional village council) and forced the communities to look for ways to “look after the village ourselves” as one farmer puts it. Under pressure from local ex-Village Development Committee and District Development Committee members the ‘back to basics’ process resulted in the adoption of democratic principles involving actual election of Mukhyias. As the same farmer continues, “we no longer make a person Mukhyia on the grounds of his wealth or influence in politics; our experience with elected Village Development Committees have made us more democratic”.

8. Current Institutional Landscapes

The processes and dynamics associated with promoting institutional plurality at local level through the engineering of new, and the re-inventing10 of old institutions may be illustrated with case of the Dhaulagiri Irrigation Development Project.11. From 1992-1996, its irrigation projects saw the formation of 66 Farmers Irrigation Associations, 43 Functional Literacy Groups and 90 Women’s Savings Groups/Mother’s Groups (under the heading of ‘Irrigation Related Income Generation’). The latter form of group is particularly illustrative of the dominant approach of the time: Savings were seen as an entry point for other income generating activities and the groups received training on everything from environmental awareness creation and vegetable nurseries to smokeless stove construction and financial management. While a pioneering player, the Dhaulagiri Irrigation Development Project was not the only one; national and international NGOs, bilateral and multilateral donors worked on related themes, with Forestry User’s Groups as the most prominent and widespread activity in natural resources management. Upon the departure of the Dhaulagiri Irrigation Development Project local NGOs, some of them erstwhile partners of the project created new alliances with new donors, and continued what they where by then good at, often in the communities where they were good at it, with ‘beneficiaries’ who had become equally good at it.

Obviously the Village Development Committees are absent in the current institutional landscape of the studied irrigated communities. However, the results of the efforts of promoting pluralism, compared to the pre-democratic monolithic patronage of the Panchayat times are evident: A total of 26 different socio-economic institutions exist across the communities, and one community boasts 13 institutions. In the hill communities the most dominant institutions are irrigation associations, savings and credit groups and mother’s groups which are found in all communities. Labour exchange systems (known as parma) are also found throughout, but with varying intensity. Drinking water groups, vegetable cooperatives, rotating credit groups and forest user committees exist in 4 out of 6 hill

10 As established in previous sections irrigation and other natural resources management institutions already existed prior to any donor intervention. 11 Under the Special Public Works Programme of the International Labour Organisation

10 communities and a selection of more specialised groups, ranging from beekeeper’s associations to seed production and coffee traders associations are found in all communities. Committees identified with specific projects (agricultural extension, drinking water) are present in projects close to the road. School and temple committees are the main socio- cultural institutions in the hills.

Relatively fewer institutions are found in the mountain communities where the customary village councils accommodate livelihood interests associated with irrigation and forestry. Second most ubiquitous are mother’s groups (which appear to handle most social, non-natural resource related conflicts) and rotating credit associations. Gompas (Buddhist monasteries) are the most common socio-cultural institutions in the mountains. In both the hills and the mountains category both the more socio-cultural and the more socio-economic institutions serve cultural and economic functions, and as social capital –defined as networks of social relations - all the listed institutions would appear to constitute assets that may affect bargaining and other forms of power, and probably levels of shared norms and identity among entire communities or among those affiliated with specific institutions.

9. Institutional Priorities and Meanings:

This intensive, often complementing input of a battery of ‘soft’ resources has had some effect on livelihoods, particularly in the hills:

“As individuals we are not recognized, so we need to establish relations. Our vegetable growing cooperative is most important because we share ideas and get information from outside. It is possible to change our living standard this way”. (Maya Devi, farmer, Lampata)

Maya Devi’s comment on the role of institutions in relation to‘recognition’ (vis-à-vis authorities, those with information, those who monopolise trade in commodities etc), the breaking of information asymmetries (through the sharing of experiences with e.g. the cultivation of new crops) and livelihoods is fairly representative in the context of institutional prioritisation. The comment, expressed in 2004, as she pointed towards her dense, well- maintained vegetable plots which her son was busy weeding, would make any promoter of local groups for poverty alleviation happy. The vegetable cooperative that she refers to has 75 member households in a locality of 122 households (70 of which are irrigated). Two other groups in the area have 9 and 14 members, respectively. Many farmers, like Maya Devi herself make Rs 60.000 (appx USD 1000) from the sale of vegetables in nearby Kusma and beyond. In line with what was the original DIDP ‘entry point’ philosophy, Maya Devi’s involvement started in 1992 when DIDP resident social mobilisers encouraged the formation of a Mother’s Group of which she was elected treasurer. Based on that group a Women’s Savings Group was formed and registered with the authorities in 1993 and Maya Devi continued as treasurer, in close interaction with the social mobiliser.

Members of the ‘entry point group’ while benefiting from the interaction with the social mobiliser received (among a number of training activities) formal ‘skills development training’, ‘off-season vegetable production training’ and Maya Devi herself, together with the chairperson of the Savings Group received financial management training. When the DIDP mobiliser left in 1996, a local NGO assisted a number of women from the Savings Group to set up a Livestock Group, and when the same NGO, in 1997, received a grant from a bilateral donor it assisted in creating the present vegetable cooperative. Most of the present members

11 of the vegetable cooperative are also members of the savings group except for a small number of households who, in the face of labour shortages, find themselves unable to adopt the rather labour intensive vegetable cultivation. Similar institutional trajectories are found in most of the irrigated communities; the Mother’s Group in particular appears to have been a successful entry point for credit provision.

Considering the success of these income generation endeavours it might have been expected that, when asked to rank institutions in terms of importance12, the vegetable and credit cooperatives would receive the highest ranking. However, Farmer’s Irrigation Associations (in the hills) and irrigation governance under the village councils (in the mountains) are considered the most important institutions by a vast majority of farmers (including the women involved with vegetable and credit cooperatives) in all communities. Village Development Committees rank second; and Mother’s Groups third followed by a plethora of institutions ranging from drinking water committees to health posts. In the following the Farmer’s Irrigation Associations and the Village Development Committees will be concentrated on:

The attachment of importance of institutions associated with irrigation goes beyond acknowledging irrigation as a basic requirement for agricultural livelihoods. The functioning of the irrigation system is strongly associated with its governance. As Apsara Sharma, secretary of the Mother’s Group in the village of Kurgha states:

“If there is no irrigation my family cannot grow enough food. If we cannot grow enough food my family gets trouble… The irrigation cannot work without the Water User’s Committee. They make rules and regulations and users have to follow them”.

Another woman farmer, the low-caste Dev Kala Bikka, also in Kurgha, when asked if the irrigation system could work without a Committee adds an equity perspective:

“Well-managed irrigation is important for our livelihoods. Without irrigation we cannot produce crops and without rules we cannot run the system, and get disputes among ourselves. If the system is managed well we can irrigate even with little water. The Committee is important for equal distribution, and without it the people who are strong and clever can benefit more than others”.

The fact that Village Development Committees rank as the 2nd most important institution in the hill communities (in 2004) may be surprising, considering that both Village Development Committees and District Development Committees (which are seen as the most important institution outside the community) were abolished in 2002. It appears, however, that despite its official burial, the remembrance of the 10-year long life of the Village Development Committee, the smallest unit of representation in the democratic sense, is very much alive on the institutional map, and that people expect it to be restored to life in the future. The high prioritisation is associated mainly with what Maya Devi, in the first quote of this section called ‘recognition’ and is fairly representative of farmer’s abilities to relate institutional priorities with wider livelihoods contexts. The following exchange, again with Maya Devi, illustrates this:

Maya Devi: Before [democracy], nobody liked to establish relations. It was difficult to mobilise even 200 rupees. Now that we have the vegetable group we have more recognition,

12 Farmers were requested to mention and rank what they considered the 6 most important institutions (sangh in nepali)

12 also the Village Development Committee and the District Development Committee recognised us. People also recognise our group. Interviewer: Why is it important to get recognition from the Village Development Committee? Maya Devi: Because a group can be registered and we received a grant. Interviewer: A grant from where? Maya Devi: From the organisations down there (points down, in the direction of the district headquarter). By forming the group, we received a grant; it came through the Village Development Committee. The Village Development Committee is an important institution because we get legal recognition from it. Interviewer: So the Village Development Committee is important because of the legal recognition? Maya Devi: Yes, there is the legal status that is why it is important, it has authority and can give the legal status. Interviewer: I still do not quite understand why legal status is important. Maya Devi: Legal means that if we want to work as a group and get benefits, we have to get a recommendation from the Village Development Committee, we have to register. We cannot coordinate work with the Village Deevelopment Committee or other organisations if we are not registered. Interviewer: Ok, I understand. Then what is the relationship with the District Development Committee? Maya Devi: The Village Development Committee does not have sufficient budget for doing much work at village level so they bring plans to the District Development Committee. For example, for the grant and also our drinking water project we went to the District Development Committee through the Village Development Committee.

This tendency to assign importance to the role of the Village Development Committee for obtaining legal status, and for getting recommendations for projects is echoed in all hill projects. While people also complain about political conflicts associated with the Village Development Committee, its abolishment is generally seen as a major problem13, as two farmers, Hari Prasad Sharma and Dhananjaya Upadhaya from Amalachaur village explain:

“After the political change [in 1990] we got the right to speak and the decentralization process gave power to the Village Development Committee. The government allocated Rs 500.000 per committee for development activities. The motor road was constructed. Various organizations were established, and there were women’s groups. This has contributed to our livelihoods. But the present conflict situation means that the Village Development Committee activities are not functioning at village level. It has become very hard to run activities without elected representative. The Village Development Committee secretary now only carries out legal work and stays in the district headquarters”

Kalabati Sherchan of Pipalbot village explains the problems in the context of irrigation:

When we had a problem the Water User’s Committee went many times to knock on the door of the Village Development Committee office, which was located in our village. They provided funds for repairing damages to the irrigation three years ago. We influenced the use of funds and their planning, and had dialogues with the chairman on how to spend funds. Now that

13 The abolishment of Village Development Committee is particularly problematic in the context of vegetable production as these were the key facilitators of contact to line agencies for farmers who have increasingly diversified their cropping portfolios away from grains and now depend on government extension services in sustaining production, particularly with respect to the provision of seeds.

13 both the chairman and the vice-chairman are absent it is no longer possible to do this. Because of the Maoist insurgency the office has also shifted to the district headquarters. Still, the Village Development Committee supports us for solving some problems, particularly the issuing of recommendation letters for line agencies

The elected Village Development Committee may be officially dead, but it continues to operate at a limited level. It actually continued to operate in many communities for quite some time after its abolishment, end even after seeking refuge in the district headquarters from Maoists, representatives continue to perform minimum functions. There is, however, no escaping the impression given by respondents that stripping the Village Development Committee and District Development Committee administrative layers off the local administrative maps left a vacuum with respect to any sense of connection that people had between the micro-setting and the state.

In many of these communities the ability to straddle the spheres between the private and the public in the given political space, in the pursuit of improved livelihoods, was enhanced by people for whom pluralistic institutional landscape meant political mobility and ‘power’, acquired through local institutions as platforms for gaining influence in the wider institutional landscape. In the hills, in particular, the connection between the village and district levels through user groups such as the Farmer’s Irrigation Associations held opportunities for power for some groups, typically male, Brahmin farmers with sizeable landholdings. These farmers, some of whom already held political office were quick to realise the advantages offered by the formal Water User’s Committees, and used these as platforms for gaining political influence, backed by their villages in their attempts at utilising political space. However, it is not only the deletion of the Village Development Committee/District Development Committee administrative layers that has severed the opportunity structures on offer during the democratic period. The presence of Maoist insurgents, and their targeting of the elite has been another factor in this respect, as Laxman Kandel, a wealthy farmer and ex-serviceman explains:

“We are punished from two sides. Since 2002 the government has not been able to implement any programmes in our village. But these days we suffer more at the hands of the Maoists. I have to pay 12 days pension per month to the Maoists. From time to time I have to donate additional funds. People who come back from abroad on leave have to pay a special tax, ranging from Rs 10.000 up to Rs 150.000 (USD 142-2150). We have to provide food and grains for free. Parents send their children to town for schooling, because the local boarding schools have been closed. Our agricultural production and sales have gone down, because the Maoists call us for frequent meetings and programmes of up to 7 days. But if we do not participate we get to be called anti-revolutionary and may be killed or expelled from the village”.

In addition to extortion, the Maoists as the de facto ‘power’ in rural Nepal (along with the armed forces), seek to control the institutional landscape. Kalabati Sherchan, the woman farmer from Pipalbot Village sums up the situation:

“No institution functions freely in our community because of the Maoist insurgency. We only have the health post, the post office and the school left. We are instructed by the Maoists not to work for the community without their approval. Maoist leaders do not like local institutions, NGOs and government agencies. They have closed the suspension bridge project, and they have declared that all women’s groups, forestry groups and other activities be

14 registered with them, otherwise they will not allow them to run. But they cannot [the Maoists] run meetings, pay visits and invite outsiders such as NGO representatives. They have already kidnapped two District Development Committee staff and kept them for a long time in their so-called labour camp. They also [the army] misbehave with us. They have arrested the forestry user committee chairperson and keep him in jail, charging him with meeting with the Maoists. He met them with a view to having two group members released from their custody”.

As already suggested, neither the decentralisation process of the 1990s, nor the more recent ‘recentralisation’ affected the irrigated communities of the mountains as government institutions, never held the same meaning as customary governance institutions. The role of the VDC was generally seen as subordinate to that of the village council and acknowledged mainly for the annual grant allocation, the use of which (always road building) was decided upon by the village headman anyway. With Maoists practically absent from the district, livelihoods are not affected by attempts at controlling institutions either.

10. Conclusion

The related economic, livelihood and resource access functions are central to the attribution of meaning to institutions in the irrigated communities of the hills. The matrices of institutions that shape livelihoods include both public and private institutions. Therefore, dichotomisation in terms of public and private institutional forms is of limited use in the analysis of livelihood construction. The farmers do not separate the spheres in practice, rather, their institutional priorities suggests that it is more important, at least in the livelihoods context of this study to focus on the spanning of spheres, i.e. the creation of micro to macro linkages. The decentralisation process created a pluralistic institutional landscape in which irrigated farmers in this particular area of the hills (but much less so in the mountains), through the spanning of institutional spheres, were able to reap benefits associated with the democratic era in terms of irrigation and other infrastructure, information, empowerment and organisation. The amputation of these linkages along with attempts on the part of the Maoists to monopolise local institutions have halted advancing livelihood trajectories associated with these benefits.

While caused by migration-led livelihoods diversification (where ‘larger’ forces associated with globalisation play the major role) that manifests itself in reduced operation and maintenance inputs and overall system performance, the amputation may also have deepened the institutional crisis faced by the irrigation institutions, owing to their subsequent inability to obtain external assistance. In this context it is interesting to note that the multi-purpose natural resource management institutions of Mustang appear better suited to deal with broader livelihood portfolios than their single-purpose counterpart institutions in the hills. The irrigation systems in Mustang are also largely unaffected, in terms of performance, by the processes of decentralisation and re-centralisation. It is therefore tempting to suggest that the elected multi-purpose institution based on inclusive logics may be more resilient than single purpose institutions based on more exclusive logics. The Mustang case, however, effective as it may be, is unique and deeply embedded in local contexts that include low population and relative socio-economic and ethnic homogeneity. For that reason, the lessons cannot be transferred elsewhere. In the hills, characterised by heterogeneity at all levels, the decentralisation and re-centralisation experience suggests that livelihoods will continue to depend on various matrices of institutions, both single-purpose and multi-purpose, inclusive and exclusive, and the extent to which decentralised political structures enable micro-macro linkages.

15 References: Berg, T.R. 2003. Democracy and rural development in Nepal: Implications for Rural Livelihoods. Unpublished paper presented at the Aalborg and Tribhuvan University workshop on gender and social change – political and cultural implications held at Aalborg University 30 September-1 October 2003. Berry, S. 1989. Access, control and use of resources in African agriculture: An introduction. Journal of the International Africa Institute. 59 (1), 1-5. Bingen, J., 2000. Institutions and sustainable livelihoods. Michigan State University (available on www.livelihoods.org). Blaikie, P., Coppard, D. 1998. Environmental change and livelihood diversification in Nepal: Where is the problem? Himalayan Research Bulletin, Vol. 2, 28-39. Cameron, J., 1998. The agricultural perspective plan: The need for debate, Himalayan Research Bulletin, 2, 11-14. Dayton-Johnson, D. 2001. Peasants and water: A review essay on the economics of locally- managed irrigation. Dalhousie University, Canada. Dietz, T., Dolsak, N., Ostrom, E., Stern P.C. 2002. The drama of the commons. In Ostrom, E., Dietz, T., Dolsak, N., Stern, P.C., Stonich, S., Weber, E.U. (Eds). The drama of the commons. National Academy Press. Ellis, F. 2000. Rural livelihoods and diversity in developing countries. Oxford University Press. Ellis, F., Biggs, S. 2001. Evolving themes in rural development 1950s–2000s. Development Policy Review, 19(4), 437-448. Overseas Development Institute. Furer-Haimendorf, C.V. 1975. Himalayan traders-life in highland Nepal. John Murray, London. GTZ, 1991. Final report on water resource utilisation inventory and potential study of Dhading District (Dhading District Development Project), Pulchowk, Lalitpur. Hardin, G.,1968. The tragedy of the commons. Science, 162, 1243-1248. KC, D., Pradhan, U. 1992. Indigenous knowledge and organisational processes: Experiences and lessons from local Nepali irrigation systems. In Gill, J., Thapa, G.B. (eds). Indigenous management of natural resources in Nepal, HMG Ministry of Agriculture/Winrock International, pp. 227-249. Leach, M., Mearns, R., Scoones, I.1997. Community-based sustainable development: Consensus or conflict. IDS Bulletin 28 (4), 1-3. Long, N. 2001. Development sociology: Actor perspectives. Routledge. Manor, J. 2004. User committees: A potentially damaging second wave of decentralisation? European Journal of Development Research, 16 (1), 192-213. Messerschmidt, D., 1995. Development studies—collection of essays. Bibliotheca Himalayica. Meynen W., Doornbos, M. 2004. Decentralising natural resource management: A recipe for sustainability and equity? European Journal of Development Research, 16 (1), 235-254. Moench, M., 2002. Local water management in a changing world – Making theory fit practice, unpublished essay prepared for IDRC (with the author). Nabli M and Nugent J.B, 1989. Analysing decentralisation policies in developing countries: a political–economy framework, Development and Change, 20.

16 Nepal Gazette, 1993. 01 March, Ch. 42, no. 46. NPC/APROSC, 1995. Agricultural perspective plan 1995, National Planning Commission/APROSC. NPC, 1992. Eighth Plan, 1992-1997, National Planning Commission. Ostrom, E., 1990. Governing the commons: The evolution of institutions for collective action. Cambridge University Press. Ostrom, E., Benjamin, P. and Shivakoti, G., 1992. Institutions, incentives and irrigation in Nepal. NISP/Workshop in Political Theory and Policy Analysis. Ostrom, E. 2002. The challenge of underperformance. In Ostrom E, Shivakoti, G, 2002 (Eds). Improving irrigation governance and management in Nepal, pp 3-33. Parajuli, U.N., Sharma, C., 2000. Study of a local water harvesting system in a micro- watershed in the Upper Mustang region of Nepal. In Banskota, M. and Chalise, S. (eds), Waters of life—Perspectives of water harvesting in the HKH. ICIMOD, , pp. 261-306. Pradhan, P., 1989. Patterns of irrigation organisation in Nepal. International Irrigation Management Institute. Pradhan, P., 2003. Farmer-managed irrigation systems (FMIS): A mode of water management, Water Nepal 9/10, (1-2), July 2001-July 2003. Pradhan, P., Yoder R., 1990. Irrigation development: The management and use of irrigation in the mountains of Nepal. ICIMOD MFS Series No. 16. Rana, J., 1992. Organisation and decision-making processes in a large farmer-managed irrigation system: The Chattis Mauja irrigation system in Nepal. In Gill, J. and Thapa, G.B. (eds), Indigenous management of natural resources in Nepal. HMG Ministry of Agriculture/Winrock International, pp. 250-264. Ribot, J. 2004. Waiting for democracy. The politics of choice in natural resource decentralisation. World Resources Institute. Sadeque, Z., 1999. Social institutions, local governance and social capital. Issues in mountain development, 99/1, ICIMOD. Scott, R.W., 2001. Institutions and organisations, Sage Publications Seddon, D. (ed) 1979. Peasants and workers in Nepal. Vikas Publishing House New Delhi. Seddon, D., Adhikari J. 2003: Conflict and Food Security in Nepal – a Preliminary Analysis. Report to Rural Reconstruction Nepal. Start, D., 2001. The rise and fall of the rural non-farm economy: Poverty impacts and policy options, Development Policy Review, 19 (4), 491-505, Overseas Development Institute. Webster, N., Engberg-Pedersen, L. 2002. In the Name of the Poor: Contesting Political Space for Poverty Reduction. Zed Books. Vermillion, D.L, 2001. Property rights and collective action in the devolution of irrigation system management in collective action. In Meinzen-Dick, R., Knox, A., Di Gregorio, M, Property fights and devolution of natural resource management-exchange of knowledge and implications for policy. CGIAR, pp. 183-220. Whelpton J. 2005. A history of Nepal. Cambridge University Press. Vinding, M., 1984. Making a living in the Nepal Himalayas: The case of the Thakalis of Mustang District. Contributions to Nepalese Studies. Journal of the Research Centre for Nepal and Asian Studies. Tribhuvan University.

17 Vinding, M. 1998. The Thakali – a Himalayan Ethnography. Serindia Publications, London. Yoder, R., 1994. Locally managed irrigation systems: Essential tasks and implications for assistance. Management transfer and turnover programs. International Irrigation Management Institute (IIMI).

18